Abstract
The prison-based N-ALIVE pilot trial had undertaken to notify the Research Ethics Committee and participants if we had reason to believe that the N-ALIVE pilot trial would not proceed to the main trial. In this paper, we describe how external data for the third year of before/after evaluation from Scotland's National Naloxone Programme, a related public health policy, were anticipated by eliciting prior opinion about the Scottish results in the month prior to their release as official statistics. We summarise how deliberations by the N-ALIVE Trial Steering-Data Monitoring Committee (TS-DMC) on N-ALIVE's own interim data, together with those on naloxone-on-release (NOR) from Scotland, led to the decision to cease randomization in the N-ALIVE pilot trial and recommend to local Principal Investigators that NOR be offered to already-randomized prisoners who had not yet been released.
Highlights
Naloxone is an opioid antagonist used for emergency resuscitation following opioid overdose
We describe how deliberations by N-ALIVE's Trial Steering-Data Monitoring Committee (TS-DMC) on N-ALIVE's own interim data, together with those on NOR from Scotland, led to the decision to cease randomization in the N-ALIVE pilot trial and to recommend to local Principal Investigators (PIs) that NOR be offered to already-randomized prisoners who had not yet been released [5]
In designing the N-ALIVE pilot trial, we had anticipated contamination between randomized groups of up to 20% because participants who had been randomized to NOR might administer their naloxone alternatively to an opioid-dependent peer who had overdosed, some of whom - unknown to us - might have been randomized to N-ALIVE's control group
Summary
Naloxone is an opioid antagonist used for emergency resuscitation following opioid overdose. With a brief history of formally eliciting prior opinion to inform the design and monitoring of RCTs funded by the UK's Medical Research Council (MRC); and some early accounts of DMC deliberations. The earliest example of formally eliciting prior opinion to inform trial design was “place your bets” about the mortality of surfactanttreated very premature babies (aged 25e29 weeks) in a RCT funded by MRC in the mid-1980s [13,14]. This “trial roulette” method was again used in the design and early stopping of the MRC's neutron therapy trial in pelvic cancer [15e18]: the minority prior belief on the relative mortality of neutrons versus photons turned out to have been consistent with trial's data. As high risk of overdose death soon after prison-release applies per-release, re-randomization was permitted provided that at least six months had elapsed since the participant's previous N-ALIVE release-date
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
